1The idea in one paragraph

Today every cell comes from another cell. But there must have been a first one, and it must have come from chemistry. The best hypothesis has stages. On the young Earth, with no oxygen and plenty of energy from lightning, ultraviolet light and volcanic heat, simple gases reacted to make carbon compounds such as amino acids; the Miller–Urey experiment showed this can happen. Some of those compounds, fatty acids, assembled themselves into tiny membrane bubbles. RNA, which can both carry a sequence and catalyse reactions, may have been the first molecule able to copy itself. Somewhere, possibly around hydrothermal vents on the seafloor, these pieces came together into cells, and one population of those early cells, LUCA, became the ancestor of everything alive today. None of it can be watched or replayed, which is why the evidence is indirect and the hypotheses are still argued over.

2Conditions on early Earth

Earth formed about 4.5 billion years ago. The planet on which life began was very different from today's, and Figure 1 sets the differences side by side.

The guide names four conditions, and each follows from the one before.

• No free oxygen. Oxygen in the air today comes almost entirely from photosynthesis, and there were no photosynthesisers yet.
• Therefore no ozone. The ozone layer is made from oxygen, so with no oxygen there was no ozone to absorb ultraviolet light, and intense UV light penetrated to the surface and the upper layers of the sea.
• Higher concentrations of carbon dioxide and methane, released by volcanoes and not yet removed by living things.
• Therefore higher temperatures, because carbon dioxide and methane are greenhouse gases.

Under these conditions, the guide says, a variety of carbon compounds may have formed spontaneously, by chemical processes that do not now occur. There was plenty of energy to drive reactions: UV light, lightning, volcanic heat. And there was nothing to destroy the products. Why does it not happen now? Two reasons. Oxygen in today's air would oxidise and break down newly formed carbon compounds. And any organic molecule that forms today is quickly eaten by bacteria. Darwin made the second point in a letter of 1871: such molecules "would be instantly devoured" now, which would not have been the case before living creatures existed.

3What makes something alive?

Before asking how life began, you need to know what you are trying to explain. Cells are the smallest units of self-sustaining life: nothing smaller can carry out all the processes of life by itself. A cell takes in energy and matter and uses them in metabolism, keeps its internal conditions stable, grows, responds, and makes more cells, all under the control of its own genes. A single bacterium does every one of these. A molecule of DNA, a ribosome or a mitochondrion taken out of a cell does none of them alone.

The functions of life you meet in A2.2 make a checklist. Figure 2 runs it against a cell and a virus.

Viruses are considered non-living, and the guide asks for the reasons.

• A virus is not a cell: it has no cytoplasm and no plasma membrane of its own making, just genetic material inside a protein coat.
• It has no metabolism: no enzymes to release energy, no ribosomes to make protein.
• It cannot reproduce by itself. It reproduces only by entering a host cell and using the host's ribosomes, enzymes and energy.
• Outside a host it does nothing at all: no growth, no response, no homeostasis. It is an inert particle.

A virus does have genes and it does evolve, which is why it sits near the boundary and why "discuss" is the right command term here. But by the standard that life is self-sustaining, it falls outside. A virus is best thought of as a set of instructions that needs a cell to be read.

4Why the first cell is hard to explain

Cells today arise only by the division of pre-existing cells; that is one of the pillars of cell theory. A cell is a highly complex structure, and no one has ever made one from simple chemicals. So explaining the first one means explaining how the same complexity could arise with no cell to make it. The guide names four things that had to evolve, and Figure 3 lines them up.

• Catalysis. The reactions of life are far too slow on their own. Some molecules had to speed up the reactions that made other molecules, so useful products could build up.
• Self-replication. Some molecule had to be able to make copies of itself, so that a successful sequence could be passed on. Without that there is no inheritance, and without inheritance there is no evolution by natural selection.
• Self-assembly. Larger structures had to form from smaller molecules without instructions, just from the molecules' own chemical properties, the way fatty acids form a membrane (section 6).
• Compartmentalisation. The chemistry had to be enclosed in a membrane, so the inside could become different from the outside, and so the products of a catalyst and a replicator stayed together instead of drifting away.

The order in which these arose is unknown, and they may have developed together. What matters is that a cell needs all four at once.

Nature of science: hypotheses must be testable. A claim in science, whether a hypothesis or a theory, has to be testable: it must be possible to make an observation or run an experiment whose result could count against it. Origin-of-life hypotheses are hard to test for two reasons the guide names. The exact conditions on pre-biotic Earth cannot be replicated: nobody knows precisely what the early atmosphere or oceans contained, and experiments can only use best guesses. And the first protocells did not fossilise: soft, microscopic bubbles of chemistry leave no trace in rock. So scientists test pieces of the story instead. Can amino acids form from simple gases? Can fatty acids form vesicles? Can RNA catalyse its own copying? Each part is testable even though the whole event can never be replayed. Say that in an answer and you have made the examiner's point.

5Evidence for the origin of carbon compounds: Miller–Urey

In 1952 Stanley Miller, a student working with Harold Urey, tested whether the conditions then thought to exist on early Earth could produce organic molecules from inorganic ones. Figure 4 shows the apparatus.

Method. A sealed glass apparatus contained a mixture of gases chosen to model the early atmosphere: methane, ammonia, hydrogen and water vapour, with no oxygen. Water in a small flask was boiled to model the ocean and to circulate water vapour. In a large flask, electric sparks between two electrodes modelled lightning. A condenser cooled the gases so that water and anything dissolved in it condensed and ran down into a trap, from which samples were taken, before the cycle continued. The apparatus ran for about a week.

Results. The water turned brown, and chemical analysis found several amino acids, the monomers of protein, along with other organic compounds. None had been present at the start. A later re-analysis of Miller's stored samples with more sensitive modern methods found a wider range of amino acids than he had first reported.

Evaluate it. "Evaluate" means weigh strengths against limitations and reach a judgement, and this is how the guide wants it examined.

Strengths.

• It was the first experimental evidence that organic molecules can form spontaneously from inorganic ones under plausible conditions, with no living thing involved.
• It turned a speculative idea into a testable one, and it has been repeated and varied many times since.
• Products were identified by chemical analysis, so the result is objective.

Limitations.

• The gas mixture may be wrong. Many geologists now think the early atmosphere was mostly carbon dioxide, nitrogen and water vapour, with less methane, ammonia and hydrogen. Such mixtures produce far fewer amino acids in the same kind of experiment.
• It made monomers, not polymers, and not cells. Amino acids are a long way from proteins, RNA or a membrane-bound cell. The experiment addresses only the first step.
• It was a small, sealed, controlled system. A real ocean is vast and dilute, so the products would be spread thinly, and the energy input in the experiment was concentrated.
• The products included both mirror-image forms of amino acids in equal amounts, whereas life uses only one form. The experiment does not explain that choice.

Judgement. The experiment shows that the first step, carbon compounds from simple gases, is chemically possible without life. It does not show how life began, and how closely it models early Earth depends on an atmosphere we cannot sample.

Other evidence supports the same first step. Some meteorites, fragments of asteroids, contain amino acids and other organic compounds formed in space, so carbon compounds may also have been delivered to the early Earth from outside, as its water may have been (A1.1).

6Membranes that build themselves

A compartment is needed so that the chemistry inside can become different from the chemistry outside. The guide's hypothesis is that the first membranes formed by themselves, from fatty acids. Figure 5 shows how.

A fatty acid has a hydrophilic head, attracted to water, and a long hydrophobic tail, repelled by it (A1.1). Put enough fatty acids in water and they arrange themselves so the tails are hidden from the water and the heads face it. One arrangement that does this is a bilayer: two layers, tails pointing inwards, heads outwards. A flat sheet of bilayer still has exposed edges, where tails meet water, so it curls round and closes into a sphere with no edges at all: a vesicle. The fatty acids coalesce into spherical bilayers spontaneously. No enzyme or gene is involved; it is the same hydrophobic effect that makes oil form droplets.

Why this matters: a vesicle is a membrane-bound compartment. Molecules trapped inside stay together; the inside can build up different concentrations from the outside. That is the first step towards a cell. In laboratory experiments, fatty-acid vesicles can take in more fatty acids and grow, and can be split into smaller vesicles by gentle forces, a physical kind of reproduction with no genes at all. Fatty-acid membranes are also more permeable than modern phospholipid membranes, which would have let small molecules such as nucleotides into early compartments.

7RNA: a presumed first genetic material

Here is the chicken-and-egg problem. In a modern cell, DNA carries the genes, but DNA cannot be copied without protein enzymes, and proteins cannot be made without the genes in DNA. Which came first? The answer many scientists favour is neither: RNA came first. Figure 6 shows why.

RNA can do both jobs.

• It can store information and be replicated. Its base sequence can be copied by complementary base pairing, just like DNA's (A1.2).
• It can catalyse reactions. A single strand of RNA can fold into a three-dimensional shape with an active site. Catalytic RNA molecules are called ribozymes, and scientists have made ribozymes in the laboratory that can join nucleotides into a copy of an RNA template, although not yet a complete copy of themselves.

So RNA may have acted at first as both the genetic material and the enzymes of the earliest cells: an "RNA world". Later, DNA, which is chemically more stable, took over long-term storage, and proteins, built from twenty different amino acids and so more versatile, took over most catalysis.

The best evidence that this happened is a relic still inside every cell. When a ribosome joins amino acids into a protein, the peptide bond is formed by ribozyme activity: the catalytic site of the ribosome is made of RNA, not protein. The most fundamental reaction in the making of every protein is still catalysed by RNA, which is what you would expect if RNA was doing that job before proteins existed.

Linking question: why RNA rather than DNA first? RNA can catalyse, DNA essentially cannot; and cells still make the building blocks of DNA by modifying the building blocks of RNA, which suggests the RNA versions came first.

8The last universal common ancestor

Whatever the first cells were like, all of today's life traces back to one population of them: the last universal common ancestor, LUCA. It was not the first life; it was the latest ancestor that every organism now alive has in common. Figure 7 shows its place.

The guide names two lines of evidence.

• The universal genetic code. Bacteria, archaea and eukaryotes all use the same codons for the same amino acids. As A1.2 explained, an arbitrary code shared by all life is best explained by inheritance from a single ancestor.
• Genes shared across all organisms. Some genes are found in every living thing, from bacteria to people, with recognisably similar base sequences: many of those for ribosome components and for the machinery of protein synthesis, for example. The simplest explanation is that they were all inherited from LUCA. By finding the genes shared across the deepest branches of the tree of life, researchers have reconstructed a picture of what LUCA's genome may have contained.

The guide also asks you to consider other forms of life. It is likely that life arose more than once, or that there were many early lineages with different chemistry. If so, why is every organism alive today descended from LUCA? The likely answer is competition: LUCA and its descendants outcompeted the others, which became extinct and left no descendants. The dashed lines in Figure 7 are those lost lineages. It follows that LUCA was not alone in its time; it was the lucky survivor.

9How dates are estimated

The guide wants you to appreciate the immense length of time over which life has been evolving, and to know how the dates are estimated. There are three approaches.

Radiometric dating of rocks. Radioactive isotopes decay at a fixed rate, measured as a half-life. By comparing how much of a parent isotope remains with how much of its decay product has built up in a mineral, the age of the rock can be calculated. If a rock contains fossils, the fossils are at least that old.

Fossils and chemical traces. The oldest widely accepted fossils of life are stromatolites: layered mounds built by mats of microorganisms, preserved in rocks in Western Australia dated to about 3.5 billion years ago. Older, more disputed evidence includes microscopic structures and chemical traces in older rocks. Living things take up the lighter isotope of carbon, carbon-12, slightly more readily than carbon-13, so carbon enriched in carbon-12 in an ancient rock is a possible chemical signature of life.

Molecular clocks. Differences in base sequence accumulate between two lineages at a roughly steady rate over time. Count the differences between the genes of two groups of organisms, calibrate the rate using fossils of known age, and you can estimate when the two lines split. Applied to the deepest branches of the tree of life, this estimates the age of LUCA. Recent estimates place LUCA at around 4 billion years ago or earlier, but the uncertainty is wide.

Figure 8 puts these dates on one line.

To feel the scale, squeeze Earth's 4.5-billion-year history into a single calendar year, starting on 1 January. Stromatolites appear in late March. Oxygen starts to build up in the air around midsummer. Animals diversify in mid-November. Modern humans appear about 35 minutes before midnight on 31 December. For most of Earth's history, life was microscopic.

10Hydrothermal vents: where LUCA may have lived

Hydrothermal vents are places on the seafloor where water heated by volcanic rock below gushes out, carrying dissolved minerals. Where the warm fluid meets cold seawater, minerals precipitate and build chimneys riddled with tiny pores. Figure 9 shows why many researchers think LUCA evolved in a setting like this.

A vent offers what section 4 said a first cell needed: a steady energy source in the chemical difference between the warm, hydrogen-rich fluid and the cold, carbon-dioxide-rich ocean; mineral catalysts, such as iron and sulfur compounds, on the chimney surfaces; and natural compartments, the pores, in which molecules could concentrate. It is also shielded from the ultraviolet light at the surface.

The guide names two kinds of evidence.

• Fossil evidence from ancient vent precipitates. Rocks in Quebec, Canada, that formed as seafloor hydrothermal vent deposits contain tiny tubes and filaments of iron oxide, similar to structures made by iron-using bacteria at vents today. The rocks are at least 3.77 billion years old and possibly much older. If the structures are fossils, they are among the oldest evidence of life, and they are in a vent setting. They are disputed, because non-living chemistry can make similar shapes.
• Conserved sequences from genomic analysis. When the genes that trace back to LUCA are reconstructed (section 8), they include genes for living without oxygen, for using hydrogen gas as an energy source, for fixing carbon dioxide, and for proteins that work at high temperatures. That is the metabolism you would expect of an organism living in a hot, hydrogen-rich vent environment.

Two different kinds of evidence, fossils and genomes, pointing to the same place is what makes the vent hypothesis strong. It remains a hypothesis: other settings, such as shallow pools on land, still have supporters.